Title
Deep feature descriptor based hierarchical dense matching for X-ray angiographic images.
Abstract
•A novel dense correspondence matching algorithm is proposed to address the limitation of matching angiographic images caused by repetitive weak-textured regions.•A deep feature descriptor, trained on angiographic images, is proposed to compute more distinctive correlation maps for correspondence matching, compared to those obtained with conventional feature descriptors.•An affine-transformation based dense completion method is further designed to improve correspondence matching accuracy from the sparse correspondence detection results.
Year
DOI
Venue
2019
10.1016/j.cmpb.2019.04.006
Computer Methods and Programs in Biomedicine
Keywords
Field
DocType
Coronary artery,Convolutional neural network,Hierarchical dense matching
Computer vision,Feature descriptor,Convolutional neural network,Image matching,Computer science,Image subtraction,Feature matching,Correlation,Artificial intelligence,Deep learning,Subtraction
Journal
Volume
ISSN
Citations 
175
0169-2607
0
PageRank 
References 
Authors
0.34
0
9
Name
Order
Citations
PageRank
Jingfan Fan15314.09
Jian Yang228348.62
Yachen Wang310.68
Siyuan Yang431.04
Danni Ai54514.78
yong huang675.23
Hong Song788.34
Yongtian Wang845673.00
Dinggang Shen97837611.27